Re: Objective-C standard

[YL]

Standard issue will be a serious one only when market acceptance getting
better,
not the other way around (Just my opinion).

And the market acceptance to me depends mainly on
(1) An IDE and runtime that works on popular OS, that easy to install,
      and can easily get started (I know it hurts but visual basic did
better on this)
(2) Rich libraries/framworks that provide solutions for business app
developers
      instead of computing genius.
(3) An app server / platform that use the language, that provides the
evidences
      of the power of the language and application examples in all major IT
aspects.

Apple's best achievement on (1) is WebObjects 4.5.1 that runs very well on
Windows
MacOS X server, and runs on various unix OS.  GnuSTEP can do better if it
can easily
run on windows. (The tech issues are solved in WO4.5)

The most efficient way of enforcing standard is to provide frameworks that
closely
reflect the business operational needs. ObjC and the existing frameworks are
the
best to me language-wise and general-computing-wise. But they can do much
better
if some more operation oriented frameworks available on top of them.
  For example EOF is very powerful but people still have to start from data
modeling
while what they did is mostly just re-invent the wheel. Plus this is
something demands
too much for newbies to make it right. Without some good examples or more
concrete
frameworks (say EObjects and EBusinessElements on top of EOF), EOF is
reasonably lack of market acceptance: it's too elegant to be used by
developers with
less OO modeling experience and full of earthly intentions.

  If a language that cannot be used to develop large scale network, data
driven
applications, many will not pay attention on it and It'll be a langue for
programming
nobles. So Gnu/ObjC version of WO/EOF is very important for the popularity
of the
language.

  The purpose of GnuSTEP, OpenGroupWare and other ObjC orgs is to provide
somethings that Apple doesn't. Without such communities outside of Apple, I
don't
see much reason why am I still using ObjC.

========================
EObjects is a framework that implements a subclass EObject of
EOGenericRecord
to obtain runtime EO class(DNA) behavior/property management (software
evolution
and evolution traceability), and two particular EOEntities: EType and
ERelation that
make the runtime/dynamic Typing and context-preserve relationship possible.

EBusinessElements is a framework that provides a generic model for business
where business Entities like Person, Participation, EnterpriseUnit, Affair,
RuleSet
etc are implemented for application to extend and override.

I'm using ObjC/WO/EOF for EObjects and EBusinessElements mainly because
there are no other language/tools can possibly work(to me:-).




----- Original Message ----- 
From: "Adrian Robert" <arobert@cogsci.ucsd.edu>
To: "DISCUSS GNUstep" <discuss-gnustep@gnu.org>
Sent: Friday, June 02, 2006 8:50 AM
Subject: Objective-C standard


> I was recently refreshing my knowledge of C++ and found myself once
> again wondering at the tremendous disparity between market acceptance
> of one over the other -- even greater than the Wintel vs. Mac(tel)
> disparity.  The lack of a standard is surely part of the problem.  If
> you use Objective-C you basically lock yourself in to a particular
> compiler (bad) which could theoretically change its language
> definition (worse).  (Of course Apple's not going to invalidate old
> code, but some of the alternatives might.)
>
> The lack of a standard library is another part of the problem, which
> the OpenStep specification and obviously GNUstep solve, but now that
> OpenStep seems basically dead GNUstep becomes something like Mono, a
> more or less precarious clone of a proprietary API that is of use
> mainly for porting existing codebases.
>
> I know Apple hosts an "ObjC language" list that might be more
> appropriate for (part of) this topic, but I'd be more curious to hear
> the perspective of (assumedly less Apple-orbiting) GNUsteppers.  And
> I guess I'm wondering whether there's anything the open source
> community can do about these issues, since Apple seems not to be
> motivated, and Sun, who helped bring OpenStep about in the first
> place and aren't always as dumb as they look ;), seems to be
> caffeinated, dripped, and filtered out of the picture.  Though I
> guess an updated OpenStep standard without Apple following it would
> be useless..  ;(
>
> Thanks for your thoughts.
>
> Adrian
>
>
>
> _______________________________________________
> Discuss-gnustep mailing list
> Discuss-gnustep@gnu.org
> http://lists.gnu.org/mailman/listinfo/discuss-gnustep
>
>